John Fruehe (JF-AMD) no longer with AMD

[georgec84]

WTF is going on with AMD...

 

[AtenRa]

Dirk was fired on Jan 10, 2011.

Bulldozer was released on Oct 12, 2011.

I'd say the Board of Directors knew how BD was (under)performing well before Jan 10.

You under-performing when you are off target. Since we dont actually know what targets they were shooting for, we cant say BD is under-performing.

If you referring to under-performing against the competition then yes BD is slower than SB. But i strongly believe that PileDriver will be a much better all around processor that will close the gap to SB and even be faster in some cases.

 

[CHADBOGA]

It wasnt uncommon for people on other forums to get bans or forced vacations if they questioned JF-AMD and his trusty followers. I got a nice permaban myself that way other places.

I also got banned for a few weeks at a time for daring to say that on the balance of probabilities, Bulldozer was going to be a dud and that IPC would be lower than Thuban.

However I think the appalling moderator in that case, may have actually been in love with JF-AMD.

While JFAMD himself was instructed from above. He did go way beyond the call of duty himself. And deserves absolutely no sympathy.

However it sure did change after BD launch and its been much more peaceful since.

His conduct was the worse I have ever seen on mainstream forums, by a company representative.

And most ludicrous of all, is how much information his supporters must be blissfully ignorant of, to not see his despicable behaviour.

AMD's engineers at a Hot Chips convention stated that one of their primary goals they were working on with Bulldozer was to minimise IPC loss, with the clear inference being that there was going to be some IPC loss.

This was publicly stated months before Bulldozer was released, yet JF-AMD kept insisting that IPC would be better than Thuban's.

 

[AtenRa]

IPC is application depended. BD has higher IPC than Phenom in some applications and worst in others.

Also, JFAMD was always commenting only for Server parts not desktop.

 

[CHADBOGA]

IPC is application depended. BD has higher IPC than Phenom in some applications and worst in others.

On the whole, any reasonable person would conclude on average it is significantly worse.

Also, JFAMD was always commenting only for Server parts not desktop.

Yeah, he would wade into discussions where participants were clearly only talking about desktop and speak that nonsense.

I wonder why this master manipulator was trying that trick.

And besides, the IPC on server was worse.

 

[AtenRa]

As i said, IPC is application depended. If your Desktop/Server workloads is mostly SIMDs, then Bulldozer IPC is way faster than previous AMD architectures.

 

[CHADBOGA]

As i said, IPC is application depended. If your Desktop/Server workloads is mostly SIMDs, then Bulldozer IPC is way faster than previous AMD architectures.

When one discusses IPC, they are talking about a reasonable average across the board, not some cherry picked niche.

You must be a disciple of JF-AMD, with all your obfuscation.

 

[Phynaz]

You under-performing when you are off target. Since we dont actually know what targets they were shooting for, we cant say BD is under-performing.

So it was intentional they designed a CPU that was slower than their previous generation.

Got it!

 

[ShintaiDK]

As i said, IPC is application depended. If your Desktop/Server workloads is mostly SIMDs, then Bulldozer IPC is way faster than previous AMD architectures.

AMD gambled on either 2 things. That software would be able to multithread in an unheard scale. And/or that they could clock it much higher.

Both was terrible wrong. And the company simply lacked any kind of foresight in the industry.

AMD also got a preference for the server segment. A segment they are totally destroyed in and will never return to. Another big mistake from the management to continue to persue that goal.

AMD=VIA.

 

[Vesku]

AMD gambled that they could create a transitional cpu for their HSA plans that would be received if not well at least acceptably. Was a pretty rough delivery. Will this refresh bring it up to acceptable levels?

It really doesn't have to trump Intel offerings, it does have to be closer in low threaded tasks to SB than Bulldozer was. Luckily for AMD, IB didn't raise the bar much but will they be able to take advantage of this?

I've mentioned this before but has AMD done any analysis to why certain tasks such as many games get such a boost on Intel designs? More so than general differences in performance would indicate. I've seen it mentioned that it could be Intel's cache system and that's very believable since cache structures seems to be where AMD runs into the most issues.

 

[AtenRa]

So it was intentional they designed a CPU that was slower than their previous generation.

Got it!

Integer

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECint®_rate2006 = 406


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECint®_rate2006 = 510

28% more performance with 33% more INT cores + 100MHz more frequency. Integer IPC is lower than previous generation but the product (Opteron 6278) is 28% faster.


Floating Point (FP)

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECfp®_rate2006 = 324


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECfp®_rate2006 = 376

Dual socket AMD Opteron 6278 has 16 FP cores (dual socket 8+8 FP cores) and yet it is 16% faster than last generation having less FP cores (16 vs 24) (with 100MHz more frequency).

Also, in desktop in most multithreaded apps the 8-core Bulldozer is faster than 6 core Phenom II. Yes it is not a great margin faster but non the less it is faster than Phenom.

The CPU as a product is not slower than last gen, Integer IPC is slower. FP IPC is way faster than last Gen.

 

[ShintaiDK]

LOL.

By that statement, SB is 100%+ faster than Nehalem and Haswell will be 100%+ faster than SB/IB.

 

[AtenRa]

When one discusses IPC, they are talking about a reasonable average across the board, not some cherry picked niche.

SIMDs and FP is used in most of multimedia applications. I dont believe that current multimedia apps are a niche in desktop. In server a lot of workloads are running in FP, and SIMDs are getting to used more and more.

I havent seen you saying that Intel's AVX and AVX-II (those are SIMDs) that everyone is touting are for a niche segment.

You must be a disciple of JF-AMD, with all your obfuscation.

When you are cornered and you cant counter anymore you taking it personal. It only shows that you dont have the knowledge to continue a technical debate.

 

[guskline]

Moderator: Time to lock this thread.

 

[Sweepr]

As i said, IPC is application depended.

So what? The overall performance per clock (common CPU intensive apps on a desktop, mixing int/FP) is lower than Phenom II. A Phenom II X6 1100T is shown to be ~25% faster than the FX6100 at same clock.

http://www.hardware.fr/medias/photos_news/00/34/IMG0034531.gif

Also, in desktop in most multithreaded apps the 8-core Bulldozer is faster than 6 core Phenom II. Yes it is not a great margin faster but non the less it is faster than Phenom.

For those expecting 50% more performance with 33% more cores, 7,5% faster in a series of MT apps sounds like a joke at best (you'd expect more than that on desktops, even if this statement was about servers).

Hopefully Piledriver will fix some of that flaws soon.

 

[exar333]

Integer

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECint®_rate2006 = 406


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECint®_rate2006 = 510

28% more performance with 33% more INT cores + 100MHz more frequency. Integer IPC is lower than previous generation but the product (Opteron 6278) is 28% faster.


Floating Point (FP)

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECfp®_rate2006 = 324


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECfp®_rate2006 = 376

Dual socket AMD Opteron 6278 has 16 FP cores (dual socket 8+8 FP cores) and yet it is 16% faster than last generation having less FP cores (16 vs 24) (with 100MHz more frequency).

Also, in desktop in most multithreaded apps the 8-core Bulldozer is faster than 6 core Phenom II. Yes it is not a great margin faster but non the less it is faster than Phenom.

The CPU as a product is not slower than last gen, Integer IPC is slower. FP IPC is way faster than last Gen.

28% more performance with 33% more cores and 100 extra mhz is FAIL for a new arch.

There is no other way to state that....sorry.

 

[jhu]

Integer

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECint®_rate2006 = 406


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECint®_rate2006 = 510

28% more performance with 33% more INT cores + 100MHz more frequency. Integer IPC is lower than previous generation but the product (Opteron 6278) is 28% faster.


Floating Point (FP)

AMD Opteron 6176 (12 Core, 2300MHz) (K10)

SPECfp®_rate2006 = 324


AMD Opteron 6278 (16 core, 2400MHz) (Bulldozer)

SPECfp®_rate2006 = 376

Dual socket AMD Opteron 6278 has 16 FP cores (dual socket 8+8 FP cores) and yet it is 16% faster than last generation having less FP cores (16 vs 24) (with 100MHz more frequency).

Also, in desktop in most multithreaded apps the 8-core Bulldozer is faster than 6 core Phenom II. Yes it is not a great margin faster but non the less it is faster than Phenom.

The CPU as a product is not slower than last gen, Integer IPC is slower. FP IPC is way faster than last Gen.

That doesn't put the Bulldozer into a better light. Per thread, Bulldozer is slower.

Per the data you posted
Integer:
Magny-Cours: 33.83 / thread
Interlagos: 31.87 / thread

FP:
Magny-Cours: 27 / thread
Interlagos: 23.5 / thread

What this shows is that AMD, for all that time and effort they put into Bulldozer, would have been better off making a dual-eight core K10h on their new 32 nm process, and it would have been faster than their K15h effort.

 

[Idontcare]

You under-performing when you are off target. Since we dont actually know what targets they were shooting for, we cant say BD is under-performing.

If you referring to under-performing against the competition then yes BD is slower than SB. But i strongly believe that PileDriver will be a much better all around processor that will close the gap to SB and even be faster in some cases.

I doubt the headcount losses at AMD in-and-around the release of Bulldozer came as a result of bulldozer performing to internal expectations.

I have no idea what bulldozer was targeted to perform at, but the fallout stemming from its actual performance certainly lends credence to the conclusion that it did in fact underperform.

Obviously John Fruehe was internally guided to expect IPC to not decrease, and yet IPC decreased, I would call that under-performing.

It underperformed internal expectations, and it underperformed external expectations that were set as a matter of corporate policy/agenda/marketing.

How much more clearly could the case be made that bulldozer underperformed and failed to reach its internal objectives?

 

[pelov]

I have no idea what bulldozer was targeted to perform at, but the fallout stemming from its actual performance certainly lends credence to the conclusion that it did in fact underperform.

They were targeting +25% over the i7 920 which would have put it within SB territory. AMD was also gunning for mid 4ghz clock speeds and IPC holding steady at Deneb/Thuban levels but they missed both goals. There's no doubt that it underperformed and I'm sure that was partly the reason for some of the delays.

The marketing and PR spin on the launch was the worst of it all, imo. They knew it underperformed and they still tried to pull whatever gimmick they could to show that wasn't the case when it quite clearly was; the overclocking record, the "blind gaming test" and the [H] answers on the forums were clearly sifted so they ultimately looked like press release statements.

 

[myocardia]

xx

 

[Conscript]

for me, it wasn't so much his info was wrong....i knew he was just a marketing guy and ultimately only had whatever information he was being fed to him. It was the manner in which he presented himself which I found annoying and unprofessional, and no amount of blame game for him "diabolically misled dis-information" as some of you would put it to absolve him of any wrongdoing, really changes the case at all. He was a jerk...it's ok to say.

 

[thilanliyan]

Dirk was fired on Jan 10, 2011.

Bulldozer was released on Oct 12, 2011.

I'd say the Board of Directors knew how BD was (under)performing well before Jan 10.

From what I have gathered, Dirk was fired for NOT having a mobile (cell phones, tablets) strategy.

 

[Idontcare]

From what I have gathered, Dirk was fired for NOT having a mobile (cell phones, tablets) strategy.

Yes :sneaky: that is clearly what it must have happened. Could not have had anything to do with faildozer, the timing was just a coincidence.

I hear it was also just coincidence that Pat Gelsinger left Intel on the eve of Intel announcing that Larrabee was being scuttled as well. The two were not related, Pat was just really really motivated to go work at EMC for personal reasons.

Had bulldozer turned out to be a hit and Dirk was let go then I'd buy the "lack of mobile strategy" cover-story. Given the facts on the ground though, no, I'm not buying it at all. Dirk gambled, and lost, on pursuing bulldozer at the expense of not pursuing alternative internally competing projects.

As captain of a ship that was steered into an iceberg, his early retirement was most certainly directly related to the performance of AMD under his stewardship and not related at all to hypothetical what-ifs that come with the "if only he had opted to develop a mobile strategy in parallel to prioritizing bulldozer".

Further proof that the mobile excuse was just that can be found in the fact that Rory has been CEO for a year now and AMD does not have anything close to a comprehensive mobile plan. If it was such a priority to the BoD as to motivate them to fire Dirk (versus any number of lesser punitive reactions) then to be sure the new CEO would have been well prep'ed to know he needed to roll out such a new mobile-facing strategy within his first 100 days of office.

But unsurprisingly Rory's AMD looks a lot like Dirk's AMD, the roadmap for bobcat and bulldozer remain in place and nothing even close to resembling a mobile strategy has taken shape, despite Rory being under the BoD's thumbs for a year now.

 

[pelov]

But unsurprisingly Rory's AMD looks a lot like Dirk's AMD, the roadmap for bobcat and bulldozer remain in place and nothing even close to resembling a mobile strategy has taken shape, despite Rory being under the BoD's thumbs for a year now.

In Read's defense, he's still new as far as the semiconductor market goes. In order to see any of his plans we'll have to wait at least 3 years to even see a single PDF slide with his influence stamped on it. Looking for "direction" from a new CEO in a market that takes 3-5 years to see a single project through is a bit silly. He had no choice but to see BD through, but he's already taken steps with regards to Steamroller. On the fab end he's cut ties to GloFo and even embraced bulk so you've certainly got to give him some credit.

 

[Rvenger]

Rory seems to be a fan of putting up the white flag and moving onto something else that hasn't sunk... yet. I really hope he has a trick up his sleeve and will show it in the next year or so. I am getting impatient.